SLAM System for Sparse Feature Environments Using Marker Matching Evaluation
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
GMapping, a Particle Filter-based SLAM algorithm, is widely used due to its robustness against odometric drift. However, in sparse feature environments such as long corridors, its localization accuracy deteriorates because of scan-matching ambiguity. To overcome this limitation, we propose an enhancement to the weight update stage of the Particle Filter pipeline. Specifically, we introduce a marker matching evaluation step that measures the consistency between observed marker data and the map from the previous frame. This enhancement enables reliable localization even in low-feature environments. Preliminary experiments demonstrate that the proposed method improves pose estimation accuracy in corridor-like indoor settings without requiring additional sensors or prior maps.
Fields of Study & Subjects
Robotics and Sensor-Based Localization
Indoor and Outdoor Localization Technologies
Inertial Sensor and Navigation
Publication details
- Journal / source
- International Conference on Artificial Intelligence in Information and Communication (ICAIIC)
- Pages
- 1274–1276
- Publication date
- 24 February 2026
- Language
- EN